Hepatitis B virus (HBV) is clearly associated with hepatocellular carcinomas, yet the mechanism of carcinogenesis is poorly understood. One possibility, strengthened by recent transgenic mouse studies, is that cellular necrosis and regeneration contribute significantly to carcinogenesis. In chronic HBV infection, there is frequent appearance of so-called ground glass cells, which are hepatocytes with intracellularly retained HBV surface proteins aggregated in the endoplasmic reticulum. Ground glass cells in transgenic mice can cause sufficient hepatocellular damage and regeneration to induce hepatocellular carcinoma. Ground glass cells are known to result from over-expression of the HBV large surface protein, but the molecular events that lead to such over-expression during chronic infection are unknown. Our hypothesis is that HBV genomic mutations that arise during chronic infection comprise one major cause of ground glass cells. During the first 3 years of this project, we have been studying in detail the cis-elements of HBV that regulate surface gene expression. As a result, we have identified regions of the HBV genome, which when mutated, can cause relative over-expression of large surface proteins and hence intracellular retention of surface proteins in cultured cells. Interestingly, these regions are frequently mutated during chronic infection. Therefore, for the proposed renewal, we wish to generate HBV clones that simulate naturally occurring HBV genomes with mutations in these regions. We will first confirm that these clones cause intracellular retention of surface proteins in cultured cells. We will then generate transgenic mice containing mutant genomes, and look for the appearance of ground glass cells, hepatitis, and hepatocellular carcinomas. It is anticipated that these studies will lead to new insights on the pathogenesis of ground glass cells, and ultimately of hepatocellular carcinoma in HBV infection.